Lepr. Rev. (1978) 49, 203-213

Absence of jj-Glucuron idase in leprae and Elevation of the Enzyme in I nfected Tissues

K. PRABHAKARAN, E. B. HARRIS AND W. F. KIRCHHEIMER

U. S. Public Health Service Hospital. Carville. LA 70 721. USA

,8-Glucuronidase actlVlty was determined in mouse footpads infected with My cobacterium leprae. in the organisms separated from the liver and spleen of experimentally infected , and in the tissues. Enzyme assays in th� mouse footpads were initiated 1 week after inoculation with M. /eprae and continued at monthly intervals for 12 months. In the mouse footpads and in the armadillo tissues, M. leprae resulted in remarkable elevations of ,8- glucuronidase leveis. The leprosy seemed to be devoid of the enzyme. In its properties like pH optimum, reaction velocity and effect of inhibitors, the activity detected in M. leprae resembled the host tis sue enzyme rather than bacterial ,8- glucuronidase; and the activity was found to be superficially adsorbed on the bacilli. lt is well established that phagocytes are rich in lysosomal enzymes. Evidently, the increased ,8-g1ucuronidase of the infected tissues is not derived from the invading organisms, but from the differenttypes of phagocytic cells infiltratingthe tissues.

Introduction j3-Glucuronidase is an important hydrolytic enzyme ubiquitously distributed in animal tissues and in tissue fluids. Phagocytic cells are especially rich in j3-g1ucuronidase. In the mammalian liver, the enzyme is largely associated with Iysosomes, and approximately one-third of the activity is distributed in the endoplasmic reticulum. The hydrolase is closely correlated with cellular pro­ liferation and tissue repair; high leveis of the enzyme are found in the reproductive and endocrine organs and in tumours. Skin cancers as well as normal skin contain j3-g1ucuronidase. The enzyme has been reported in molluscs and in insects. Among , Escherichia co li, Streptococcus pyogenes, xerose, C. hofJmani and rumen micro-organisms have been shown to produce j3-g1ucuronidase (Fishman, 1955; Levy and Marsh, 1959). In bacterial characterized by intracellular parasitism of the invading organisms, marked e1evation of hydrolytic enzymes has been observed (AlIen, 1969). j3-Glucuronidase splits offD- glucuronic acid residues from oligosaccharides, thus degrading these complex molecules. In the body, glucuronic acid forms

Received for publication 9 February 1978.

0305-75 18/78/090 1-0203$01 .00/0 IC> 1978 British Leprosy Relief Association 204 K. PRABHAKARAN ET AL. conjugates of hormones, drugs and other substances, thus serving a detoxifica­ tion function; also ,B-glucuronidase can hydrolyze such conjugates and release the organic compounds (Fishman, 1955). Because of their degradative activities, Iysosomal enzymes, including ,B-glucuronidase, have a major role in the killing and digestion of ingested bacteria (Allen, 1969). In leprosy, as irt certain other mycobacterial infections, a paradoxical situation occurs; the bacteria, instead of being destroyed by the phagocytes, seem to be protected in these cells (Goren, 1977; Brown et aI., 1969). No significant elevation of ,B-glucuronidase was observed in peripheral blood leucocytes of leprosy patients (Avila and Convit, 1970; Garcia­ Gonzalez et ai., 1977). Increase in acid phosphatase was demonstrated histochemically in histiocytes containing My cobacterium leprae (Brieger and Allen, 1962). Based on histochemical evidence, it was concluded that M. leprae possesses ,B-glucuronidase (Matsuo and Skinsnes, 1974). In the course of our investigations over the past several years on Iysosomal enzymes in leprosy, we studied the leveis of acid phosphatase and ,B-glucuronidase in mouse footpads and in tis sues of armadillos experimentally infected with M. leprae. We also determined ,B-glucuronidase activity in concentrates of the bacilli separated fr om infected armadillo tis sues. Because acid phosphatase is extremely labile, its assay was found to be unre\iable in tissues stored for varying periods of time. As such, acid phosphatase assays were not pursued fu rther. However, ,B-glucuronidase gave consistent and reproducible results. The data presented in this report show that M. leprae purified fr om infected armadillo tissues apparently does not contain ,B-glucuronidase; in the mouse fo otpad, the enzyme activity continues to rise as the infection progresses; and in the infected armadillo tis sues, the enzyme leveis are elevated 2 to 3 times over that of the uninfected tissues.

Materiais and Methods

ANIMALS ANO MYCOBACTERIA Approximately 8-week-old female Swiss mice ofthe NIH strain were used in the study. The experimental animais were inoculated in the left hind footpads with 1 x 104 M. leprae. The M. leprae used was a mouse-passage strain originally obtained fr om the skin biopsy of an untreated lepromatous patient. Control mice were of the same age and sexo Starting at 1 week after inoculation, a fe w mice (usually 5) were sacrificed at monthly intervals for 12 months. The soft tissues from the left hind footpads of both the experimental and the control animals were collected and stored at -80°C. At the time of assay, the material was thawed out, washed in cold saline, blotted and minced. The tissues were extracted in small amounts of water by grinding in a chilled agate mortar. The extract was centrifuged briefly, to remove the particulate niatter. Enzyme determinations were made in the supernatant fr action. was estimated by the method of Lowry et aI. (195 1). Enumeration of M. leprae in the mouse footpad was done at 6 months and 12 months, by the method of Hanks et ai. (1964), modified by Kirchheimer. Liver and spleen of "normal" and infected armadillos were coUected GLUCURONIDASE AND LEPROSY 205

aseptically at autopsy and were held at O°C or at -80°C before processing. Ten percent homogenates of these organs were made in water. Suspensions of the leprosy bacilli were prepared as reported before (Prabhakaran et aI., 1976) from the liver or spleen of 9-banded armadillos experimentally infected with M. leprae (Kirchheimer and Storrs, 1971; Kirchheimer et ai., 1972). The bacilli were disrupted by ultrasonic oscillation (Prabhakaran et aI., 1973). In an effort to remove superficially adsorbed host-tissue materiais from the bacterial concentrate, it was treated with trypsin (final concentration 0.5%), N aOH (0. 1 N), sodium dodecyl sulfate (1%), or sodium deoxycholate (0.5%) for 30 min at o°C, except for trypsin, which was done at 37°C . My cobacterium phlei was grown in Proskauer-Beck medium (Youman's modification) at 37°C for 2 weeks. The bacilli were harvested by centrifuga­ tion and washed twice with cold saline and once with deionized glass-distilled water.

ENZYMES AND CHEMICALS Purified E. coZi ,lj-glucuronidase (Type VII), bovine liver ,lj-glucuronidase (Type B-IO), phenolphthalein glucuronic acid (sodium salt) and glycine buffer were purchased fr om the Sigma Chemical Co., St Louis, MO, USA. Other chemicals used were ofthe highest purity commercially available.

ENZYME ASSAY The enzyme assay was done essentially by the standard procedure of Talalay et aI. (1946). The reaction mixture consisted of the following con­ stituents: 0. 1 M acetate buffer (pH 4.7), 0.4 ml; 0.01 M phenolphthalein glu­ curonic acid, 0. 1 ml; sample, 0.5 ml. The sample blank contained 0.5 ml buffer and 0.5 ml sample, and the reagent blank 0.9 ml buffer and 0.1 ml substrate. For E. coZi ,lj-glucuronidase, 0.075 M phosphate buffer (pH 6.8) was substituted for acetate buffer; 40 units of the enzyme were added. The mycobacterial suspensions used contained 2-5 x 109 organisms/ml. The final concentration of the inhibitors tested was 0.005 M. After thereaction mixture was incubated at 37°C for 60 min, 1 ml of 5% TCA was added, followed by 2.5 ml of alkaline glycine reagent. Volume was made up to 6 ml with 1.5 ml of water. The reaction mixture was centrifuged and the purple colour of the supernatant fr action was read at 540 nm in a Beckman DU-2 spectrophotometer. The readings were corrected for any absorbance due to the sample or reagent blanks. The amount of phenolphthalein liberated from the substrate by the enzyme was calculated from a standard curve. All reactions were carried out in duplicate and each experiment was done at least 3 times. The results reported are mean values of representative experiments.

Results

MOUSE FOOTPADS M. leprae inoculated in the mouse footpads multiplied normally; enumeration of the bacilli showed 2.4 ± 0.18 x 106 organisms per footpad in 6 206 K. PRABHAKARAN ET AL.

60

50 O i;;c e o- O> 40 E ...... c � o .c 30 1:: Cl. o c ., .c 20 o- O' ::L

10

O 2 4 6 8 10 12 Time (months)

Fig. I. ,B-glucuronidase of mouse footpads infected with M. leprae. (Á-Á) Experimental; (e-e) control.

months, and 1.8±O.15x 107 in 12 months. j3-Glucuronidase actlVlty was assayed in the footpad tissues at monthly intervals for 12 months. To correct for variations in individual assays, the values obtained for both the control and the experimental samples were subjected to regression analysis. The results are presented in Fig. 1. In the uninfected mice, the enzyme activity remained fairly stable throughout the experimento However, j3-glucuronidase continued to rise in the footpads of mice where M. leprae multiplied progressively. These results do not indicate whether the higher activity is derived from the bacilli or from the cells that infiltrate the infected footpads.

ARMADILLO TISSUES j3-Glucuronidase was determined in the liver and the spleen tissues of both the infected and the uninfected armadillos. (On the average, the infected organs

TABLE 1 ,B-Glucuronidase in "normal" and irifected sp leen tissue of armadi/los: Ilg of phenolphthalein released/mg protein

Number of animals "Normal" Infected

1 32.2 92.1 2 37.7 173.5 3 44.4 110.5

Mean 38.1 125.4 GLUCURONIDASE ANO LEPROSY 207

TABLE 2 f3-Glucuronidase in "normal" and infected liver tissue of armadillos: I1g of phenolphthalein releasedlmg protein

N umber of animais "Normal" Infected

1 77.4 188.9 2 162.0 200.0 3 118.4 286.3 4 130.8 234.4

Mean 122.2 227.4

contained over 109 M. leprae per g of tissue.) Compared to the controls, the activity in the infected spleen was elevated by over 3 times, and by almost 2 times in the liver (Tables 1 and 2).

M. LEPRAE It was not clear whether the source of the elevated enzyme activity observed was the invading organisms or the host tissues. To ascertain this, M. leprae was s�parated from both the spleen and the liver tissues and f3-glucuronidase activities determined in the bacterial preparations. The preparative procedure involved difTerential and density-gradient centrifugations in solutions of sucrose and KCI (Prabhakaran et ai., 1976). When assayed without further purification, these bacterial suspensions contained f3-glucuronidase. When the bacilli were disrupted by ultrasonic oscillation, more enzyme acitivity was not released. Both intact and disrupted organisms gave similar results. When the disrupted suspension was fr actionated, the supernatant contained a greater amount of f3-glucuronidase than the particulate fraction. Additional washing resulted in further loss of enzyme activity from the bacterial particles (Table 3). Evidently the enzyme is not firrnly attached to the bacterial membranes.

TABLE 3 Distribution of f3-glucuronidase in M. leprae disrupted by ultrasonic oscillation: I1g of phenolphthalein released

Sample Unheated Heated

U nfractionated suspension 47.0 o Particulate fr action 15.0 O Supernatant fraction 20.0 O I wash of particulate fraction 5.0 O

To determine whether the actlvlty was due to enzyme molecules superficially adsorbed on the bacilli,intact M. leprae preparations were treated with trypsin, NaOH and 2 detergents. Trypsin-treatment removed over 50% of the activity from the bacilli. NaOH and sodium dodecylsulfate produced total inactivation, and only a residual activity was left on treatment with deoxycholate (Table 4). Probably, f3-glucuronidase is not an inherent property of the leprosy bacilli, but is only superficially attached to the organisms. 208 K. PRABHAKARAN ET AL.

TABLE 4 E.lJe ct of dijJerent'treatments on {J-glucuronidase of M. leprae: Ilg of phenolphthalein released

Treatment Untreated Treated bacilli bacilli

Ultrasonic oscillation 29.0 28.0 Trypsin 66.0 37.0 Deoxycholate 46.0 7.0 Sodium dodecyl sulfate 46.0 O NaOH 65.0 O

TABLE 5 {J-Glucuronidase in M. phlei: Ilgof phenolphthalein released

Sample Unheated Heated

Intact bacilli O O Disrupted bacilli O O

Another mycobacterium tested, M. ph/ei, both intact and disrupted had no (3- glucuronidase (Table 5). pH OPTIMA In order to characterize the enzyme activity detected in the untreated M. leprae suspensions, properties of the enzyme were compared with those of (3- glucuronidase of E. calí and of mammalian liver. Bacterial and mammalian (3- glucuronidases are distinguished from one another by their pH optima. With phenolphthalein glucuronide as substrate, optimal hydrolysis by the mammalian enzyme takes place at pH 4.5-5.2, and by the bacterial enzyme at pH 6- 7 (Levy and Marsh, 1959). The pH-activity curve of the enzyme in armadillo liver is shown in Fig. 2. The pH optima for the {j-glucuronidase of E. calí and of the M. leprae suspension are given in Fig. 3. It may be seen that the activity detected in M. leprae resembles c1ose1y that of the host tis sue, rather than the bacterial enzyme.

REACTION VELOCITY A T HIGH SUBSTRATE LEVELS Increasing substrate concentrations resulted in a pronounced inhibitory efTecton E. calí (3-glucuronidase. No such efTectwas detected with mammalian (3-glucuronidase or with M. leprae; the results obtained with armadillo liver and with the leprosy bacilli gave hyperbolic curves. The reaction velocity of the liver enzyme is shown in Fig. 4. The results obtained with E. calí and M. leprae are presented in Fig. 5. It is evident that the activity in the leprosy bacilli is similar to that of the host tissue enzyme. Figure 6 represents the double-reciprocal Lineweaver-Burk plot of the reaction velocity of E. calí (3-glucuronidase. Km value of the enzyme was 5 fo und to be 8.0 x 10- M. This compares favourably with the Km of 7.5 x 1O-5M reported for sheep rumen bacteria (Levy and Marsh, 1959), GLUCURONIDASE ANO LEPROSY 209

200 I- /0 "- o o

c 'o; 150 I- O "O / \ .<::: O L' a. (5 c l Q) O .<::: 0- 100 \0 O' ::l.. I \ 50 O 0 \ 0 O/ "" O 2,0 30 40 50 60 70 80 pH

Fig. 2. pH-Activity curve: armadillo tiver.

x 40 l- \ x c 'o; 30 - "O .<::: L' a. o 1"'- \ c Q) x .<::: o- 20 / O' ::l.. \.

1 0 - o "- o x �o / /x \0 i J J I O 20 30 40 50! 60 70 \80 90 pH ,. Fig. 3. pH-Activity curve: M. leprae and E. coli. (e--e)M. leprae; (x-x) E. coli 210 K. PRABHAKARAN ET AL.

I I I I I o 2 4 6 8 10 4 Substrate concentration (x 1() M ) Fig. 4. Reaction velocity at high substrate leveIs: armadillo liver. 50 /� 40 I . c . "(jj � "O � .c 30 a. .� o c Q) .c D- .�. O> 20 ::l... �x

10 x xI O 2 4 6 8 10 4 Substrate concentration (x 10- M) Fig. 5. Reaction velocity at high substrate leveIs: M. leprae and E. coli. ()C-)C)M. leprae; (e---e) E. coli. GLUCURONIDASE AND LEPROSY 211

7

6

- 2 5 -2 0 - 1 ,5 - 1 ,0 -0,5 O 0 5 1 , 0 1,5 2 ,0 3,0 1/ [5] (XI0 4 M)

Fig. 6. E. co/i tl-glucuronidase: double-reciprocal plot of substrate concentration versus reaction velocity. confirming the reliability of the assay procedure we adopted for (3- glucuronidase. However, it may not be valid to derive such kinetic data with crude preparations like tis sue homogenates and the M. leprae suspensions.

INHIBITORS In addition to the above criteria, bacterial and mammalian (3-glucuronidases are difTerentiated by the efTects of inhibitors on the enzymes. Contrary to what has been reported recently (Matsuo et aI., 1975), ascorbic acid by itself has no inhibitory efTect on (3-glucoronidase (Levy and Marsh, 1959). We tested 2 specific inhibitors of the enzyme M. leprae suspensions, armadillo liver homogenates, and purified bovine liver and E. colí (3-glucuronidases. The results are given in Table 6. The mammalian enzyme is completely inhibited by

TABLE 6 E./fe ct of inhibitors on {J-glucuronidase:Ilg of phenolphthalein released

D-Saccharic acid- l ,4-lactone Galactosaccharic acid Sample % % -Inhibitor +Inhibitor Inhibition -Inhibitor + Inhibitor Inhibition

Armadillo liver 115 O 100 99 26 74 M. leprae 32 O 100 37 5 87 tl--glucuronidase (beef li ver) 192 O 100 110 11 90 tl--glucuronidase (E. coli) 56 14 75 59 59 O 212 K. PRABHAKARAN ETAL.

D-saccharic acid- l ,4-lactone; the E. calí enzyme is inhibited only partially. Galactosaccharic acid has no efTect on E. calí, whereas it shows 70-90% inhibition of the mammalian enzyme. The activity present in M. leprae behaves the same way as that in the mammalian liver.

Discussion The data presented here demonstrate that in mouse fo otpads as well as in armadillo tissues, M. leprae infection results in a significant elevation of (3- glucuronidase activity. This increased activity is apparently not derived from the infecting organism; probably it originates in the phagocytic cells which infiltrate the tis sues. Phagocytes are known to be rich in Iysosomal enzymes incIuding (3-glucuronidase. Infiltration of armadillo tissues infected with M. leprae by difTerent types of phagocytic cells has been reported (Kirchheimer et ai., 1972; Job et aI., 1978). It has been shown that the mouse fo otpad where M. leprae multiples is infiltrated by polymorrhonucIear cells, Iymphocytes and monocytes (Evans et ai., 1973). The activity detected in the M. leprae separated fr om the armadillo tissues seems to be of host-tissue origino In its properties like pH optima, reaction velocity and efTect of inhibitors, the activity in the bacilli resembled (3- glucuronidase derived from mammalian tissues rather than the bacterial enzyme. Moreover, the enzyme seemed to be superficially adsorbed on the organisms, and could be easily removed by various treatments. Apparently, (3- glucuronidase is not an intrinsic property of M. leprae. This concIusion is supported by the finding that another Iysosomal enzyme, acid phosphatase, detected in in viva-grown tubercIe bacilli is derived from the host cells (Kanai, 1969). The precise role of the Iysosomal acid hydrolases in leprosy infection remains to be elucidated.

References

AlIen, J. M. (1969). Lysosomes in bacterial infection. Lysosomes in Biology and Medicine, Vol. 2, pp. 41-68. J. T. Dingle and H. B. Fell, Eds. North-Holland Publishing Co., Amsterdam. Avila, J. L. and Convit, J. (1970). Studies on cellular immunity in leprosy. I. lysosomal enzymes. Int. J. Lepr. 38, 359. Brieger, E. M. and Allen, J. M. (1962). Cytopathological changes in lepra cells. Expt Ce/l Res. 28, 438. Brown, C. A., Draper, P. and D'Arcy Hart, P. (1969) Mycobacteria and Iysosomes: a paradoxo Nature (Lond.) 221, 658. Evans, M. J ., Newton, H. E. and Levy, L. (1973). Early response of mouse footpads to My cobacterium leprae. bifect. Im mun. 7, 76. Fishman, W. H. (1955). p-Glucuronidase. Advances in Enzymol. 16, 361. Garcia-Gonzalez, J. E., Rojas-Espinosa, O. and Estrada-Parra, S. (1977). Phagocytosis in leprosy. I. The leveis of diaphorase, P-glucuronidase, acid phosphatase and lipase in circulating leukocytes. Lepr. Rev. 48, 17. Goren, M. B. (1977). Phagocyte Iysosomes: interactions with infectious agents, phagosomes, and experimental perturbations in function. A nn. Rev. M icrobiol. 31, 507. Hanks, J. H., Chatterjee, B. R. and Lechat, M. F. (1964). A guide to the counting of mycobacteria in clinicaiand experimental materiais. Int. J . Lepr. 32, 156. GLUCURONIDASE AND LEPROSY 213

Job, C. K., Kirchheimer, W. F. and Sanchez, R. M. (1978). Liver lesions in experimental lepromatoid leprosy ofthe armadillo--ahi stopathological study. In!. J. Lepr. 46, I. Kanai, K. (1967). Detection of host-originated acid phosphatase on the surface ofin vivo-grown tuberc1e bacilli. Jap . J. Med. Sei. Biol. 20, 73. Kirchheimer, W. F. and Storrs, E. E. (1971). Attempts to establish the armadillo (D asypus novemcinetus Linn.) as a model for the study ofleprosy. I. Report oflepromatoid leprosy in an experimentally infected armadillo. Int. J. Lepr. 39, 692. Kirchheimer, W. F., Storrs, E. E. and Binford, C. H. (1972). Idem. 11. Histopathologic and bacteriologic post-mortem findingsin lepromatoid leprosy in the armadillo. In!. J. Lepr. 40, 229. Levy, G. A. and Marsh, C. A. (1959). Preparation and properties of j3-g1ucuronidase. Advan. Carbohydr. Chem. 14, 381. Lowry, O. H., Rosenbrough, N. J., Farr,A. L. and Randall, R. J. (1951). Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265. Matsuo, E. and Skinsnes, O. K. (1974). Acid mucopolysaccharide in leprosy. 2. Subcellular localization of hyaluronic acid and j3-glucuronidase in leprous infiltrates suggestive of a host-Myeobaeterium leprae metabolic relationship. Int. J. Lepr. 42, 399. Matsuo, É. and Skinsnes, O. K. (1974). Acid mucopolysaccharide metabolism in leprosy. 2. mycobacterial gtowth enhancement, and growth suppression by saccharic acid and vitamin C as inhibitors of j3-g1ucuronidase. Int. J. Lepr. 43, I.

Prabhakaran, K., Harris, E. B. and Kirchheimer, W. F. (1973). Particulate nature of 0- diphenoloxidase in My eobaeterium leprae and assay of the enzyme by the radioisotope technique. Mierobios 8, 151. ­ Prabhakaran, K., Harris, E. B. and Kirchheimer, W. F. (1976). Hypopigmentation of skin lesions in leprosy and occurrence of o-diphenoloixdase in Myeobaeterium leprae. Pigment Ce/l. Vol. 3, pp. 152-164. V. Riley, Ed. S. Karger, Basel. Talalay, P., Fishman, W. H. and Huggins, C. (1946). Chromogenic substrates. 11. Phenolphthalein glucuronic acid as substrate for the assay of glucuronidase activity. J. Biol. Chem. 166, 757.